Du et al.

BMC Complementary and Alternative Medicine 2014, 14:326

http://www.biomedcentral.com/1472-6882/14/326

RESEARCH ARTICLE Open Access

Antidepressant-like effects of the hydroalcoholic

extracts of Hemerocallis Citrina and its potential
active components
Bingjian Du, Xiaoshuang Tang, Fei Liu, Chunyue Zhang, Guanghua Zhao*, Fazheng Ren and Xiaojing Leng*

Abstract
Background: Herbal therapies are potential alternatives and adjuncts for depression treatment. The present study
aims to investigate the antidepressant-like effects of hydroalcoholic Hemerocallis citrina extracts and its potential
neuropharmacological components.
Methods: Hydroalcoholic H. citrina extracts were phytochemically analyzed. Behavioral models, including tail
suspension tests and open field tests, were performed to evaluate the antidepressant-like effects of the extracts. A
possible mechanism was explored by analyzing brain monoamine neurotransmitters. Toxicity and histopathological
analyses were performed to determine whether or not the extracts are safe for oral administration.
Results: The antidepressant-like effects of hydroalcoholic H. citrina extracts were mainly related to flavonoids, especially
rutin and hesperidin. The extract prepared using 75% ethanol (i.e., HCE75) exhibited the highest active flavonoid content
and activity. Orally administered 400 mg/kg of HCE75 significantly induced an antidepressant-like effect, whereas the
combination of equivalent rutin and hesperidin dosages exhibited the same profiles. Isobologram analysis
showed sub-additive antidepressant interactions between rutin and hesperidin. HCE75 (400 mg/kg, p.o.) increased
the serotonin and dopamine levels in the central nervous system. Mortality and lesions were not observed upon
oral administration of up to 5000 mg/kg HCE75.
Conclusions: The antidepressant-like effects of hydroalcoholic H. citrina extracts are mainly related to flavonoids,
especially rutin and hesperidin. The serotonergic and dopaminergic systems may have major roles. The active
extract is toxicologically safe for oral administration.
Keywords: Hemerocallis citrina, Antidepressant-like effects, Neurotransmission, Flavonoids

Background essential. Psychotropic plants that exhibit multiple bio-

Depression is a chronic, relapsing, and potentially fatal
disorder that affects about 20% of the population. De-
pression has been projected to become the second most
common disorder worldwide by 2020 [1]. Antidepressant
drugs often have undesirable side effects, such as cholin-
ergic symptoms, withdrawal issues, sexual dysfunction
[2], and worsening insomnia [3]. Therefore, developing
effective depression therapies with few side effects is
* Correspondence: gzhao1000@163.com; lengxiaojingcau@163.com
CAU & ACC Joint-Laboratory of Space Food, College of Food Science &
Nutritional Engineering, Key Laboratory of Functional Dairy Science of Beijing
and Ministry of Education, Beijing Higher Institution Engineering Research
Center of Animal Product, Beijing Dairy Industry Innovation Team, China
Agricultural University, No.17 Qinghua East Road, Haidian, Beijing 100083,
People’s Republic of China
activity and few side effects have gained significant at-
tention as complementary and alternative medicines
[4,5].
Emerging evidence suggests that plant flavonoids
mainly exert beneficial effects on the central nervous
system (CNS) by protecting neurons against stress-
induced injury, suppressing microglia and astrocyte acti-
vation and promoting synaptic plasticity, memory, and
cognitive function [6,7]. These properties depend on the
chemical structure of the flavonoids. Schinus molle L.
(Anacardiaceae) and Hypericum species are rich in rutin
and have been reported to be antidepressants mediated
by their interaction with serotonergic, noradrenergic,
and dopaminergic systems [8]. The antidepressant-like

© 2014 Du et al.; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative
Commons Attribution License (http://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and
reproduction in any medium, provided the original work is properly credited. The Creative Commons Public Domain
Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article,
unless otherwise stated.
Du et al. BMC Complementary and Alternative Medicine 2014, 14:326
http://www.biomedcentral.com/1472-6882/14/326
effect of hesperidin on mice depends on its interaction
with serotonergic 5-HT1A receptors [9].
Hemerocallis citrina (daylily) is a plant widely grown
in East Asia that has antibacterial [10], antioxidant [11],
and nitrite-eliminating activities [12]. H. citrina has
been clinically efficient in relieving depression in pa-
tients aged 11 to 80 years [13]. A previous study
showed that the ethanol extract of Hemerocallis fulva
has an antidepressant-like effect, in which rutin is be-
lieved to have an important role [14]. The ethanol ex-
tract of H. citrina has been recently reported to elicit
antidepressant-like effects depending on monoaminer-
gic systems [15]. Some researchers have also suggested
that such activity of the ethanol extract is at least partly
mediated by neurotrophic [16] and inflammation sys-
tems [17]. However, the relationships between specific
H. citrina neuropharmacological activities and its fla-
vonoid components remain uninvestigated. The scien-
tific evaluation of its antidepressant effects are still not
convincing.
In this study, the relationships between the H. citrina fla-
vonoid composition and its corresponding antidepressant-
like activities were determined by performing tail
suspension tests, open field tests, and neurochemical
analyses in mice. Toxicity and histopathological ana-
lyses were also carefully conducted.
Methods
Animals
Male imprinting control region mice (25 g to 36 g in
weight) were purchased from Vital River Laboratories
(Beijing, PR China). The mice were housed in cages and
were given unrestricted access to food and water at 25 ±
1°C. Humidity was controlled at 56% ± 3% in a room
maintained on a 12 h light/dark cycle (lights on at 8 a.
m.). The mice were randomly assigned into specified ex-
perimental groups (n = 10 animals per group) and were
utilized only once. All animal procedures were con-
ducted in accordance with the animal care and use
guidelines of the China Council on Animal Care (Regu-
lations for the Administration of Affairs Concerning Ex-
perimental Animals approved by Decree No. 2 of the
State Science and Technology Commission on Novem-
ber 14, 1988). The experiments were approved by the
Animal Experimental Welfare and Ethical Inspection
Committee, The Supervision, Inspection, and Testing
Center of Genetically Modified Organisms, Ministry of
Agriculture (Beijing, China).
Plant materials and preparation of extracts
Dried H. citrina flowers (7.3% ± 1.2% moisture content;
n = 6) were purchased from Wal-Mart (Beijing, PR China).
A 100 g botanical sample of the flowers was finely pow-
dered and extracted through maceration three times at
Page 2 of 11
25 ± 2°C for 12 h with 1 L portions of deionized water
containing different ethanol concentrations (i.e., 0%, 25%,
50%, 75%, and 100%). Ethanol was selected as solvent to
ensure food safety. The H. citrina extracts (HCEs) were
filtered and lyophilized at −60°C for 48 h. Freeze-dried ex-
tracts were produced using deionized water containing 0%
(HCE0), 25% (HCE25), 50% (HCE50), 75% (HCE75), and
100% ethanol (HCE100). The HCEs were sealed and
stored in a freezer before use.
Drugs and treatments
D-glucose, bovine serum albumin, quercetin, quercitrin,
isoquercitrin, rutin, hyperoside, hesperidin, serotonin (5-
HT), norepinephrine (NE), and dopamine (DA) were
provided by Sigma-Aldrich (St. Louis, MO, USA). Fluox-
etine, sodium pentobarbital, and diazepam were pur-
chased from China National Medicines (Beijing, PR
China). All chemicals and reagents were analytical grade
unless otherwise stated.
All experiments were performed between 13:00 and
17:00. Different groups of mice were used for each test.
The HCEs were dissolved in physiological saline with 2%
Tween 80; the other drugs were dissolved in physio-
logical saline immediately before use. The reagents were
orally administered (p.o.) at the constant volume of
10 ml/kg body weight. The control group mice received
appropriate vehicles.
The mice subjected to the experiment for the
antidepressant-like effects of HCE received behavioral
tests 60 min after they were administered (p.o.) with
the vehicle (physiological saline with 2% Tween 80),
fluoxetine (20 mg/kg), and HCE (400 mg/kg). The mice
were tested 60 min after they were administered (p.o.)
with either the vehicle (physiological saline with 2%
Tween 80), 20 mg/kg fluoxetine, 0.1, 1, 2, 4, and 8 mg/
kg rutin, 0.03, 0.3, 1, 2, and 4 mg/kg hesperidin, 0.01,
0.1, 0.2, 0.4, and 0.8 mg/kg quercetin, and 0.01, 0.1, 0.2,
0.4, and 0.8 mg/kg quercitrin to investigate their anti-
depressant effects. The mice underwent behavioral tests
0, 0.5, 1, 2, 3, 4, and 8 h after administering (p.o.) with
the vehicle (physiological saline with 10% Tween 80),
HCE75 (400 mg/kg), standardized flavonoid mixture
(8 mg/kg), and (75:21.5, w/w) rutin/hesperidin fixed-
ratio combination (8 mg/kg) to investigate the temporal
evolutions of the antidepressant-like effects of the treat-
ments. The mice underwent behavioral tests 1 h after ad-
ministration (p.o.) of the vehicle (physiological saline with
10% Tween 80), fluoxetine (20 mg/kg), HCE75 (20, 200,
400, 800, and 1600 mg/kg), standardized flavonoid mix-
tures (0.4, 4, 8, 16, and 32 mg/kg), and (75:21.5, w/w) rutin
and hesperidin (0.4, 4, 8, 16, and 32 mg/kg) to investigate
their antidepressant-like effects.
The mice of the groups that were not subjected to be-
havioral tests were decapitated and subjected to brain
Du et al. BMC Complementary and Alternative Medicine 2014, 14:326 Page 3 of 11
http://www.biomedcentral.com/1472-6882/14/326

surgery 66 min after administration (p.o.) of the vehicle Table 1 Chemical composition of HCE (%, w/w)
(physiological saline with 2% Tween 80), fluoxetine Sample Yield Carbohydrate Protein Crude fat Flavonoid
(20 mg/kg), and HCE75 (20, 200, 400, 800, and 1600 mg/ HCE0 18.2 ± 1.2 80.1 ± 4.1 9.2 ± 0.6 0.3 ± 0.1 0.4 ± 0.1
kg) to investigate the effects of HCE75 on monoamine HCE25 16.9 ± 0.9 81.8 ± 3.3 8.5 ± 0.4 1.1 ± 0.2 0.6 ± 0.1
neurotransmitter levels in prefrontal cortex and hippo-
HCE50 14.9 ± 0.8 75.2 ± 2.9 8.1 ± 1.0 1.7 ± 0.3 0.7 ± 0.1
campus of mice without or with behavioral tests. The mice
of the blank group were not treated, but similarly under- HCE75 9.2 ± 0.6 69.4 ± 5.3 6.7 ± 0.7 6.3 ± 0.7 2.0 ± 0.2
went brain surgery. The mice of the groups exposed to the HCE100 4.4 ± 0.5 11.4 ± 0.8 0.2 ± 0.1 74.1 ± 2.1 2.3 ± 0.2
behavioral tests were decapitated and underwent brain Data are expressed as mean ± S.E.M. (n = 3).
surgery immediately after the animals received the tests
60 min after the administration (p.o.) of the samples. performed using a Purospher STAR RP-18 column
(150 mm × 4.6 mm, 5 μm). Solvents A (acetic acid–water,
Tail suspension test (TST) 5:95) and B (pure methanol) were programmed at 0 (95%
TST is a behavioral model widely used to assess A + 5% B), 17 (88.5% A + 11.5% B), 28 (88.3% A + 11.7%
antidepressant-like activities by measuring the mobility B), 57 (28.3%A + 71.7% B), and 60 min (95% A + 5% B), at
effects in each test. The immobility behavior displayed 0.8 ml/min flow rate for chromatography. The column
in rodents subjected to unavoidable and inescapable temperature was maintained at 40°C, the injection volume
stresses during TST reflects behavioral despair, which re- was 10 μl, and the detection wavelength was 254 nm. The
flects depression in humans [18]. The test was per- flavonoid compounds were identified by comparing the
formed according to previously described methods [19]. retention time with authentic standards, and quantitative
In a typical procedure, acoustically and visually isolated analysis was performed in triplicate to obtain external cali-
mice were suspended 20 cm above the floor using adhe- bration curves.
sive tape placed approximately 1 cm from the tip of their
tails. The total duration of immobility during a 6 min Isobologram analysis
period was recorded. The mice were considered immo- Synergy was graphically assessed through isobologram
bile only when they passively hung or stayed completely analysis [27]. In brief, the median effective doses of the
motionless. The video was analyzed using Smart 3.0.02 drugs and their combinations were determined at a fixed
(Panlab, USA). ratio; dose–response curves were plotted in rectangular
coordinates. The straight line connecting the points of
Open-field test (OFT) the drugs represents their theoretical additive combin-
False-positive results can be obtained in TST for agents ation. The effects of the drugs are purely additive (no
with psychostimulant effects that could be measured interaction) when the point of their combination and
through OFT [20]. The mice were evaluated in an open- their S.E.M. lie on this line. Super-additivity exists when
field paradigm, as previously described [21]. The mice the points lie below this line, but sub-additivity (antag-
were individually placed in a Plexiglass box (40 cm × onism) exists when the points lie above this line.
40 cm × 50 cm) divided into 16 squares. Squares crossed
with all paws were considered as indicators of locomotor
activity. The behavioral parameters were recorded for
6 min. The floor of the open-field apparatus was cleaned
between tests.

Phytochemical analysis of HCE

The total carbohydrate content was measured by phe-
nol–sulfuric acid methods using D-glucose as a standard
[22]. Protein content was measured by Bradford assay,
using bovine serum albumin as a standard [23]. The
crude fat content was measured using the method
920.39 from the Association of Official Analytical
Chemists [24]. The total flavonoid content was assessed
through the aluminum chloride colorimetric method
and expressed as quercetin equivalents [25]. Flavonoid
profile analyses were conducted using a high-performance Figure 1 Flavonoid fingerprint of HCE75 as a representative
HPLC chromatogram. The peaks correspond to (1) rutin, (2)
liquid chromatography (HPLC) system based on previous
hesperidin, (3) quercitrin, and (4) quercetin.
studies [26]. Qualitative and quantitative analyses were
Du et al. BMC Complementary and Alternative Medicine 2014, 14:326 Page 4 of 11
http://www.biomedcentral.com/1472-6882/14/326

Table 2 Composition of the HCE flavonoids (%, w/w)

Sample Quercetin Quercitrin Isoquercitrin Rutin Hyperoside Hesperidin
HCE0 0.8 ± 0.1 0.6 ± 0.1 ND 96.2 ± 5.7 ND 0.2 ± 0.0
HCE25 5.9 ± 1.0 3.8 ± 0.2 ND 73.2 ± 1.9 ND 15.2 ± 2.1
HCE50 3.7 ± 0.9 1.3 ± 0.4 ND 65.7 ± 8.4 ND 28.1 ± 5.3
HCE75 1.5 ± 0.4 2.1 ± 0.3 ND 74.2 ± 6.8 ND 21.3 ± 3.7
HCE100 3.3 ± 0.7 3.7 ± 0.5 ND 69.2 ± 5.9 ND 24.1 ± 2.4
Data are expressed as mean ± S.E.M. (n = 3); ND, not determined.

Monoamine neurotransmitter levels
The 5-HT, NA, and DA concentrations in the mouse
brain were measured through HPLC with electrochem-
ical detection, as previously described [28]. The mice
were decapitated, and their brains were immediately re-
moved. The frontal cortex and hippocampus were care-
fully dissected and stored at −80°C until measurement
(no more than 1 week). The brain tissues were homoge-
nized in an ice-cold 0.4 M perchloric acid (5 μl/mg) solu-
tion that contains 5 mM sodium bisulfate and 0.04 mM
EDTA to avoid oxidation. The homogenate was centri-
fuged at 15,000 × g for 15 min at 4°C. Approximately 10 μl
of the resulting supernatant was chromatographed on a
Purospher STAR RP-18 column (150 mm × 4.6 mm,

Figure 2 Effects of HCE on immobility time in TST (A) and line crossings in OFT (B). The apparent correlations between (C) chemical and
(D) flavonoid constituent dosages and the immobility time in HCE in the TST. The results are expressed as% immobility time relative to the
control group (vehicle) for the antidepressant-like effects. The mice were tested 60 min after they were administered (p.o.) with the vehicle
(physiological saline with 2% Tween 80), fluoxetine (20 mg/kg), and HCE (400 mg/kg). Data are expressed as mean ± S.E.M. (n = 10). Data
were analyzed using one-way ANOVA for multiple comparisons, followed by post hoc Student–Newman–Keuls test. *p < 0.05, **p < 0.01, and
***p < 0.001 compared with the control group (vehicle); and #p < 0.05, ##p < 0.01, and ###p < 0.001 compared with the positive reference
group (fluoxetine). Linear regression was performed along with the ordinary least squares estimation.
Du et al. BMC Complementary and Alternative Medicine 2014, 14:326 Page 5 of 11
http://www.biomedcentral.com/1472-6882/14/326

5 μm). Separation was performed in an isocratic elution
mode using a mobile phase at 30°C column temperature.
The phase comprised 8% methanol and 92% water, con-
taining EDTA (0.5 mM), triethylamine (5 mM), sodium 1-
heptanesulfonate (0.5 mM), Na2HPO4°12H2O (20 mM),
and citrate (50 mM), at 1 ml/min flow rate. The neuro-
transmitters were measured using the electrode potential
of a glassy carbon working electrode at +650 mV against
the Ag/AgCl electrode. The 5-HT, NA, and DA were
identified and quantified by comparing their retention
times and peak areas with those of standard solutions;
their contents were expressed as ng/g wet weight tissue.
Acute toxicity and histopathological analysis
Orally acute HCE toxicity was estimated according to
the procedure reported by Lorke [29]. Each of the three
different animal groups (n = 3 animals per group) were
administered (p.o.) with 10, 100, and 1000 mg/kg of ex-
tract during the first phase. The doses were then in-
creased to 1600, 2900, and 5000 mg/kg in another three
different groups (n = 3 animals per group) when mortal-
ity was not observed during the first day of the first
phase and 1 week thereafter. The mice were observed
for 14 d to note possible mortality.
Histopathological analysis was conducted in a separate
series of experiments. Mice (n = 10 animals per group)
were orally treated with 5000 mg/kg HCE75 once a day
for 1 d or 21 consecutive days, and the mice were sacri-
ficed for histopathological analyses according to the previ-
ously described methods [30]. Livers, kidneys, and spleens
were immediately collected and placed in formalin. The
samples were transferred to a cassette and immersed in
multiple baths containing progressively higher volumes of
ethanol to dehydrate the tissue, followed by xylene and

Figure 3 Effects of the pure rutin (A), hesperidin (B), quercetin (C), and quercitrin (D) on immobility time in TST. The mice were tested
60 min after they were administered (p.o.) with the vehicle (physiological saline with 2% Tween 80), 20 mg/kg fluoxetine, 0.1, 1, 2, 4, and 8 mg/kg
rutin, 0.03, 0.3, 1, 2, and 4 mg/kg hesperidin, 0.01, 0.1, 0.2, 0.4, and 0.8 mg/kg quercetin, and 0.01, 0.1, 0.2, 0.4, and 0.8 mg/kg quercitrin. Data are
expressed as mean ± S.E.M. (n = 10). Data were analyzed using one-way ANOVA for multiple comparisons, followed by post hoc Student–New-
man–Keuls test. *p < 0.05, **p < 0.01, and ***p < 0.001 compared with the control group (vehicle); #p < 0.05, ##p < 0.01, and ###p < 0.001 compared
with the positive reference group (fluoxetine).
Du et al. BMC Complementary and Alternative Medicine 2014, 14:326
http://www.biomedcentral.com/1472-6882/14/326
hot paraffin. More paraffin was added during embedding
to create a paraffin block, which allowed the sectioning of
the tissues into 2 μm slices. The microtome slices were
stained with hematoxylin–eosin stain.
Statistical analysis
All values were expressed as mean ± S.E.M. The data
were analyzed using ANOVA, followed by post hoc Stu-
dent–Newman–Keuls test for multiple comparisons. Sta-
tistically significant differences were denoted by p < 0.05.
Linear regression was performed along with the ordinary
least squares estimation.
Results and discussion
Phytochemical profile analysis of HCE
The HCE0, HCE25, HCE50, HCE75, and HCE100 extrac-
tion yields and their main compositions are shown in
Table 1. Increasing ethanol concentration in the extracting
solvents decreased the carbohydrate and protein contents
and increased the crude fat and flavonoid contents in the
extract. The increase in crude fat content in HCE100 indi-
cates the enhanced fat-soluble compound extraction by
pure ethanol. Nevertheless, the extracted fat-soluble com-
pounds prevent the further extraction of the flavonoids
[31]. The extracted flavonoid in HCE100 did not rapidly
increase.
The flavonoid components of HCE were analyzed
through HPLC. Figure 1 illustrates the representative
chromatogram of HCE. Peaks 1, 2, 3, and 4 correspond
to rutin (52.6 min), hesperidin (53.2 min), quercitrin
(54.8 min), and quercetin (57.9 min), respectively. Rutin
is the major component of HCE, followed by hesperidin.
The quercitrin and quercetin contents were relatively
minor. Table 2 lists the main flavonoid components.
Antidepressant-like effects of HCE and
neuropharmacological component analysis
The effects of HCE treatment (400 mg/kg, p.o.) on im-
mobility time in TST are presented in Figure 2A.
HCE50 (p < 0.05), HCE75 (p < 0.001), and HCE100 (p <
0.01) significantly decrease immobility time compared
with the control group, indicating good antidepressant-
like performances. The reducing abilities of HCE75
were closest among the three to that of fluoxetine,
which is the most common clinical therapy drug used
as positive reference. Significant variations in OFT line
crossings were not found in Figure 2B, indicating that
HCE did not stimulate locomotor activity, thus the
antidepressant-like effects of HCE in the TST were spe-
cific. The poor anti-immobility correlations with the
carbohydrates (R2 = 0.16), proteins (R2 = 0.13), and
crude fat (R2 = 0.03), unlike flavonoids (R2 = 0.92), de-
rived from their dose–response regression analyses
shown in Figure 2C indicate a possible disconnect with
Page 6 of 11
any antidepressant-like effect [32]. Quercetin (R2 =
0.05) and quercitrin (R2 = 0.08) were also ruled out in
the regression analyses because of the apparent noncor-
relation between the flavonoids and antidepressant-like
performance (Figure 2D). By contrast, the high degree
of correlation between immobility time and dose indi-
cates that the antidepressant-like effects are closely as-
sociated with rutin (R2 = 0.96) and hesperidin (R2 =
0.97).
Pure flavonoids were tested to confirm the hypothesis.
Figure 3A and 3B show that pure rutin (2–8 mg/kg) and
hesperidin (1–4 mg/kg) elicits significant antidepressant-
like actions on mice in TST, similar to that evoked by
the equivalent HCE50 (2.1 and 0.7 mg/kg rutin and hes-
peridin, respectively), HCE75 (6.7 and 2.3 mg/kg rutin
and hesperidin, respectively), and HCE100 (5.9 and
1.9 mg/kg rutin and hesperidin, respectively). The orally
administered quercetin and quercitrin doses did not ex-
hibit antidepressant-like effects in the present conditions
(Figure 3C and 3D). Traces of contributions from carbo-
hydrates, proteins, and crude fat were not observed. Psy-
chostimulant effects were likewise not observed in the
OFT of the flavonoids (data not shown). Dimpfel [33]
found that 5 to 80 mg/kg rutin doses produced electro-
pharmacograms similar to those of clinical antidepres-
sants in rats. A single i.p. hesperidin administration
yielded antidepressant-like effects [9].
Figure 4 Temporal evolutions of the antidepressant-like effects
caused by HCE75, the standardized flavonoid mixture
containing equivalent doses of rutin, hesperidin, quercetin, and
quercitrin and the fixed-ratio combination of rutin and
hesperidin (75:21.5, w/w) on mice in TST. The mice were tested
0, 0.5, 1, 2, 3, 4, and 8 h after administering (p.o.) with the vehicle
(physiological saline with 10% Tween 80), HCE75 (400 mg/kg),
standardized flavonoid mixture (8 mg/kg), and rutin/hesperidin
fixed-ratio combination (8 mg/kg). Data are expressed as mean ± S.E.M.
(n = 10). Data were analyzed using one-way ANOVA for multiple
comparisons, followed by post hoc Student–Newman–Keuls test.
*p < 0.05, **p < 0.01, and ***p < 0.001 compared with the control group
(tested 1 h after vehicle treatment).
Du et al. BMC Complementary and Alternative Medicine 2014, 14:326 Page 7 of 11
http://www.biomedcentral.com/1472-6882/14/326

Antidepressant-like effect of HCE75 and active flavonoid
involvement
The previous analysis reveals that HCE75 is the most ac-
tive H. citrina extract. The temporal behavior of the
antidepressant-like effects and the dose-dependent prop-
erties (100 mg/kg to 1600 mg/kg) of the extract was esti-
mated. The standardized flavonoid mixture prepared
using four types of pure flavonoids and the combination
with pure rutin and hesperidin at a fixed ratio (75:21.5, w/
w) with respect to the real flavonoid composition of
HCE75 (75% rutin, 21.5% hesperidin, 2% quercetin, and
1.5% quercitrin, w/w) were compared. Figure 4 shows the
comparison of the three samples. The antidepressant-like
effects of HCE75, standardized flavonoid mixture, and
rutin/hesperidin fixed-ratio combination reached their
maxima (p < 0.001) at 1, 0.5, and 0.5 h, respectively. The
anti-depressant effects remained significant (p < 0.01) up
to 4 h after p.o. administration. The temporal evolution
curves were consistent. Psychostimulant effects were not
observed in OFT (data not shown). Therefore, the func-
tionality of the major antidepressants in HCE75 is medi-
ated by rutin and hesperidin, whereas the contributions of
quercetin and quercitrin are insignificant.
HCE75 (200–800 mg/kg) exhibited a significant anti-
immobility effect (p < 0.01) in TST compared with the
control groups (Figure 5A). Doses more than 800 mg/kg
increased immobility time and formed a U-shaped
dose–response profile in antidepressant-like action. In-
significant variations in the OFT line crossings (Figure 5B)
ruled out the psychostimulant and sedative effects. The
phenomenon was often observed during rodent model
screening (i.p. or p.o.) on potential antidepressants, which

Figure 5 Effects of HCE75, the standardized flavonoid mixture containing equivalent rutin, hesperidin, quercetin, and quercitrin doses
and the fixed-ratio combination of rutin and hesperidin (75:21.5, w/w), on (A) immobility time in TST and (B to D) their respective line
crossings in OFT. The mice were tested 1 h after administration (p.o.) of the vehicle (physiological saline with 10% Tween 80), fluoxetine
(20 mg/kg), HCE75 (20, 200, 400, 800, and 1600 mg/kg), standardized flavonoid mixtures (0.4, 4, 8, 16, and 32 mg/kg), and the fixed-ratio rutin and
hesperidin combination (0.4, 4, 8, 16, and 32 mg/kg). Data are expressed as mean ± S.E.M. (n = 10). Data were analyzed using one-way ANOVA for
multiple comparisons, followed by post hoc Student–Newman–Keuls test. *p < 0.05, **p < 0.01, and ***p < 0.001 compared with the control group
(vehicle); ##p < 0.01 and ###p < 0.001 compared with the positive reference group (fluoxetine).
Du et al. BMC Complementary and Alternative Medicine 2014, 14:326
http://www.biomedcentral.com/1472-6882/14/326
was attributed to either the activation of different path-
ways at diverse doses [34] or the suppression of the max-
imum response in the presence of antagonists through
non-competitive antagonism [4]. The dose-dependent di-
versity of the mechanisms of the neuropharmacological ef-
fects should be studied further because the highest
HCE75 dosage (i.e., 1600 mg/kg) for oral treatment re-
sulted in a hypnotic response (data not shown). The
response might negate the antidepressant-like effects.
The standardized flavonoid mixture containing equiva-
lent rutin, hesperidin, quercetin, and quercitrin doses
and the fixed-ratio combination of rutin and hesperidin
induced the same characteristics and changes without
significantly changing the OFT line crossings (Figure 5C
and 5D). These results indicate that rutin and hesperidin
cause the antidepressant effects of HCE75. The contri-
butions of quercetin and quercitrin in the effects were
insignificant.
Dose–response curves were obtained for rutin, hesperi-
din, and their fixed-ratio combination (75:21.5, w/w;
Figure 6A) to assess the interaction between rutin and
hesperidin in the antidepressant-like effects in TST. The
data were fitted to logistic dose–response curves via non-
linear regression. Dose-dependent antidepressant-like ef-
fects were observed for the chosen dose ranges with
2.5 mg/kg rutin, 1.2 mg/kg hesperidin, and 3.7 mg/kg
fixed-ratio rutin/hesperidin combination median effective
dose values. The isobolograms obtained for the rutin/hes-
peridin fixed-ratio combination (Figure 6B) reveal a
sub-additive effect. The result implies an antagonistic
interaction between rutin and hesperidin. Few studies
focused on the assessment of the neuroactive properties
of flavonoid-flavonoid interactions. The active potency
of a given flavonoid is closely linked to its structure,
Figure 6 Dose–response curves of the immobility time in TST with respect to the control groups. (A) The positive reference group
(fluoxetine, 20 mg/kg, p.o.) response was 0 for the antidepressant-like effects of rutin, hesperidin, and the fixed-ratio combination of rutin/hesperidin
(75:21.5, w/w). (B) The consequent isobolograms. Data are expressed as mean ± S.E.M. (n = 10).
Page 8 of 11
such as the ortho-dihydroxy structure in the B-ring, the
2-3-double bond in conjugation with a 4-oxo function,
and the presence of the 3- and 5-OH functions [35]. More
detailed studies are required to elucidate the mechanisms
involved in the neuropharmacological interactions.
Effects of HCE75 on monoamine neurotransmitter levels
in mice brains
Depression symptoms are associated with changes in
monoamine neurotransmitter (i.e., NE, DA, and 5-HT)
levels in the CNS [1,36]. The prefrontal cortex and hippo-
campus, which regulate emotion, motivation, learning,
and memory, are essential in depression [37]. Nutt [36] re-
vealed that NE is related to alertness, energy, and atten-
tion, and that DA is linked to pleasure, reward, and
motivation in life. The 5-HT transmitter is related to com-
pulsion, obsession, and anxiety.
Figure 7 shows the monoamine neurotransmitter
levels (i.e., NE, DA, and 5-HT) in the prefrontal cortex
(Figure 7A and 7C) and hippocampus (Figure 7B and
7D) of the mice without (Figure 7A and 7B) or with
TST (Figure 7C and 7D), respectively, after treatment.
The DA level in the hippocampus was too low to be
detected.
Insignificant differences were observed in the monoamine
neurotransmitter levels in the prefrontal cortex (Figure 7A)
or hippocampus (Figure 7B) of the blank, vehicle, fluoxet-
ine, and HCE75 treatment groups without TST. The NE
levels of HCE75-treated groups with TST were almost simi-
lar to that of the vehicle, and the levels significantly (p <
0.05) decreased after exposure to the behavioral models.
Anti-reduction was observed in the fluoxetine-treated
group (Figure 7D). DA and 5-HT levels were significantly
higher (p < 0.05) with TST and the HCE75 dose of more
Du et al. BMC Complementary and Alternative Medicine 2014, 14:326 Page 9 of 11
http://www.biomedcentral.com/1472-6882/14/326

Figure 7 Effects of HCE75 on monoamine neurotransmitter levels in the prefrontal cortex (A and C) and hippocampus (B and D) of
mice without (A and B) or with TST (C and D). The mice of the groups unexposed to TST were decapitated and were subjected to brain
surgery 66 min after administration (p.o.) of the vehicle (physiological saline with 2% Tween 80), fluoxetine (20 mg/kg), and HCE75 (20, 200, 400,
800, and 1600 mg/kg). The mice of the blank group were untreated, but similarly underwent brain surgery. The mice of the groups exposed to
TST were decapitated and were subjected to brain surgery immediately after the animals were tested 60 min after treatment. Data are expressed
as mean ± S.E.M. (n = 10). Data were analyzed using one-way ANOVA for multiple comparisons, followed by post hoc Student–Newman–Keuls
test. xp < 0.05, yp < 0.01, and zp < 0.001 compared with the blank group; ap < 0.05, bp < 0.01, and cp < 0.001 compared with the vehicle groups
exposed to TST.

than 200 mg/kg than the vehicle groups. These levels
reached their maxima at 400 mg/kg in almost all cases, and
were maintained for doses up to 1600 mg/kg, the highest
dose tested in the present work. This point is consistent
with the anti-immobility effects of HCE75 described in the
previous section. This condition indicates that increasing
the 5-HT and DA levels in the CNS is an important param-
eter in the mechanism of the antidepressant-like response
of HCE75. Machado et al. [8] proved the involvement of
5-HT, DA, and/or NE systems in the antidepressant-like
action of rutin. The antidepressant-like performance of
hesperidin was determined to be related to the 5-HT sys-
tem, whereas NE and DA systems were indirectly involved
[9], consistent with the findings in this study. However,
the potential interaction between rutin and hesperidin
requires further study. The U-shaped dose–response pro-
file observed in the previous section should not be due to
the decrease in antidepressant effect of HCE75, because
the increasing effects of HCE75 at doses ranging from
400 mg/kg to 1600 mg/kg on the monoamine neurotrans-
mitter levels in the TST were basically stable.
Toxicity and histopathological analysis
Plants rich in neuroactive flavonoids are usually inedible,
which makes the consideration of safety and tolerability
a requirement in plant supplement development. The
toxicity study on HCE75 by Lorke [29] revealed that the
median lethal dose is not feasibly estimated because
mortality was not recorded from the oral administration
of 5000 mg/kg HCE75.
Du et al. BMC Complementary and Alternative Medicine 2014, 14:326
http://www.biomedcentral.com/1472-6882/14/326
Figure 8 Representative histopathological sections of the (A) liver, (B) kidney, and (C) spleen of the mice treated with HCE75
(5000 mg/kg, p.o.) once a day for 1 d (acute treatment) or 21 consecutive days (chronic treatment). The control group was not treated.
Insignificant lesions were observed in the experiments. Hematoxylin–eosin stains, 400 × original magnification.
Figure 8 shows that the histopathological examination
has insignificant differences. The results reveal that HCE75
(5000 mg/kg, p.o.) administration once a day either for 1 d
or 21 consecutive days does not cause lesions on the exam-
ined physiological organs. Oral HCE75 administration per
se could thus be considered toxicologically safe.
Conclusions
In summary, the present TST, OFT, and neurochemical
analyses of the monoamine neurotransmitters in mice
brain revealed the significant antidepressant effects of H.
citrina. The effect of HCE75, the most active hydroalco-
holic extract of H. citrina, is dose-dependent. The results
confirm that such antidepressant effect is mainly related
to the contributions of flavonoids, especially rutin and
hesperidin. Isobologram analysis showed the sub-additive
interaction between rutin and hesperidin. Toxicity and
histopathological analyses confirmed that HCE75 is toxi-
cologically safe for oral administration.
Competing interests
The authors declare that they have no competing interests.
Authors’ contributions
BD designed the study, performed experiments, analyzed results and
contributed to writing of the paper. XT, FL and CZ helped to perform
experiments. GZ, FR and XL contributed to scientific discussion and edited
Page 10 of 11
manuscript. All authors read, edited and ultimately approved the final
manuscript.
Acknowledgments
This research was supported by the National Natural Science Foundation of
China (No. 31171771), the National Science and Technology Support
Program (2011BAD23B04), and Beijing Dairy Industry Innovation Team.
Received: 25 April 2014 Accepted: 29 August 2014
Published: 1 September 2014
References
1. Krishnan V, Nestler EJ: The molecular neurobiology of depression. Nature
2008, 455(7215):894–902.
2. Papakostas G: Tolerability of modern antidepressants. J Clin Psychiatr 2007,
69:8–13.
3. Pack AI, Pien GW: Update on sleep and its disorders. Annu Rev Med 2011,
62:447–460.
4. Huerta-Reyes M, Herrera-Ruiz M, González-Cortazar M, Zamilpa A, León E,
Reyes-Chilpa R, Aguilar-Rojas A, Tortoriello J: Neuropharmacological in vivo
effects and phytochemical profile of the extract from the aerial parts of
Heteropterys brachiata (L.) DC. (Malpighiaceae). J Ethnopharmacol 2013,
146(1):311–317.
5. McClatchey WC, Mahady GB, Bennett BC, Shiels L, Savo V: Ethnobotany as a
pharmacological research tool and recent developments in CNS-active
natural products from ethnobotanical sources. Pharmacol Ther 2009,
123(2):239–254.
6. Spencer J: Flavonoids: modulators of brain function. Br J Nutr 2008,
99(Suppl 1):ES60–ES77.
7. Jäger AK, Saaby L: Flavonoids and the CNS. Molecules 2011,
16(2):1471–1485.
8. Machado DG, Bettio LE, Cunha MP, Santos AR, Pizzolatti MG, Brighente I,
Rodrigues ALS: Antidepressant-like effect of rutin isolated from the
Du et al. BMC Complementary and Alternative Medicine 2014, 14:326 Page 11 of 11
http://www.biomedcentral.com/1472-6882/14/326

ethanolic extract from Schinus molle L. in mice: Evidence for the 31. Xu Q, Shen Y, Wang H, Zhang N, Xu S, Zhang L: Application of response
involvement of the serotonergic and noradrenergic systems. Eur J surface methodology to optimise extraction of flavonoids from fructus
Pharmacol 2008, 587(1):163–168. sophorae. Food Chem 2013, 138(4):2122–2129.
9. Souza LC, de Gomes MG, Goes AT, Del Fabbro L, Boeira SP, Jesse CR: 32. Yang H, Lee Y-C, Han K-S, Singh H, Yoon M, Park J-H, Cho C-W, Cho S: Green
Evidence for the involvement of the serotonergic 5-HT1A receptors in and gold kiwifruit peel ethanol extracts potentiate pentobarbital-induced
the antidepressant-like effect caused by hesperidin in mice. Prog Neuro- sleep in mice via a GABAergic mechanism. Food Chem 2013, 136(1):160–163.
Psychopharmacol Biol Psychiatry 2013, 40:103–109. 33. Dimpfel W: Rat electropharmacograms of the flavonoids rutin and
10. Zhan L, Li G, Li S, Wu S: Study on Extraction of Flavonids in Hemerocallis quercetin in comparison to those of moclobemide and clinically used
Citrina Baroni and Its Medical Effects. J Nanhua Univ (Medical Edition) reference drugs suggest antidepressive and/or neuroprotective action.
2005, 1:036. Phytomedicine 2009, 16(4):287–294.
11. Lang N, Luo H: Study for the activity of flavonoid in Hemerocallis citrina. 34. El-Alfy AT, Ivey K, Robinson K, Ahmed S, Radwan M, Slade D, Khan I, ElSohly
Food Res Dev 2007, 28:74–77. M, Ross S: Antidepressant-like effect of Δ9-tetrahydrocannabinol and
12. Fu M, Chen Q, Mao L: Suppression of nitrosation by extracts of nightlily other cannabinoids isolated from Cannabis sativa L. Pharmacol Biochem
(Hemerocallis citrina) flowers. Food Sci 2009, 30:114–119. Behav 2010, 95(4):434–442.
13. Chen G, Zhang C, Fan Y, Chen J: Treatment of 156 cases of depression 35. Hidalgo M, Sánchez-Moreno C, de Pascual-Teresa S: Flavonoid–flavonoid
related insomnia by Hemerocallis citrina. Zhejiang J Traditional Chinese interaction and its effect on their antioxidant activity. Food Chem 2010,
Med 2008, 43:397. 121(3):691–696.
14. Lin S, Chang H, Chen P, Hsieh C, Su K, Sheen L: The Antidepressant-like 36. Nutt D: Relationship of neurotransmitters to the symptoms of major
Effect of Ethanol Extract of Daylily Flowers (Jīn Zhēn Huā) in Rats. J Tradit depressive disorder. J Clin Psychiatr 2008, 69:4.
Complement Med 2013, 3(1):53. 37. Mao Q, Xian Y, Ip S, Che C: Involvement of serotonergic system in the
15. Gu L, Liu Y, Wang Y, Yi L: Role for monoaminergic systems in the antidepressant-like effect of piperine. Prog Neuro-Psychopharmacol Biol
antidepressant-like effect of ethanol extracts from Hemerocallis citrina. Psychiatry 2011, 35(4):1144–1147.
J Ethnopharmacol 2012, 139(3):780–787.
16. Yi L, Li J, Li H, Zhou Y, Su B, Yang K, Jiang M, Zhang Y: Ethanol extracts from doi:10.1186/1472-6882-14-326
Hemerocallis citrina attenuate the decreases of brain-derived neurotrophic Cite this article as: Du et al.: Antidepressant-like effects of the
factor. TrkB levels in rat induced by corticosterone administration. hydroalcoholic extracts of Hemerocallis Citrina and its potential active
J Ethnopharmacol 2012, 144(2):328–334. components. BMC Complementary and Alternative Medicine 2014 14:326.
17. Liu X-L, Luo L, Liu B-B, Li J, Geng D, Liu Q, Yi L-T: Ethanol extracts from
Hemerocallis citrina attenuate the upregulation of proinflammatory
cytokines and indoleamine 2, 3-dioxygenase in rats. J Ethnopharmacol
2014, 153(2):484–490.
18. Cryan JF, Mombereau C, Vassout A: The tail suspension test as a model for
assessing antidepressant activity: review of pharmacological and genetic
studies in mice. Neurosci Biobehav R 2005, 29(4):571–625.
19. Colla AR, Machado DG, Bettio LE, Colla G, Magina MD, Brighente I, Rodrigues
ALS: Involvement of monoaminergic systems in the antidepressant-like
effect of Eugenia brasiliensis Lam. (Myrtaceae) in the tail suspension test in
mice. J Ethnopharmacol 2012, 143(2):720–731.
20. Machado DG, Cunha MP, Neis VB, Balen GO, Colla A, Bettio LE, Oliveira Á, Pazini
FL, Dalmarco JB, Simionatto EL: Antidepressant-like effects of fractions,
essential oil, carnosol and betulinic acid isolated from Rosmarinus officinalis
L. Food Chem 2013, 136(2):999–1005.
21. Rodrigues ALS, Rocha JBT, Mello CF, Souza DO: Effect of Perinatal Lead
Exposure on Rat Behaviour in Open‐Field and Two‐Wky Avoidance Tasks.
Pharmacol Toxicol 1996, 79(3):150–156.
22. Dubois M, Gilles KA, Hamilton JK, Rebers PT, Smith F: Colorimetric method
for determination of sugars and related substances. Anal Chem 1956,
28(3):350–356.
23. Bradford MM: A rapid and sensitive method for the quantitation of
microgram quantities of protein utilizing the principle of protein-dye
binding. Anal Biochem 1976, 72(1):248–254.
24. Kim Y, Mosier NS, Hendrickson R, Ezeji T, Blaschek H, Dien B, Cotta M, Dale
B, Ladisch MR: Composition of corn dry-grind ethanol by-products: DDGS,
wet cake, and thin stillage. Bioresour Technol 2008, 99(12):5165–5176.
25. Chang C, Yang M, Wen H, Chern J: Estimation of total flavonoid content
in propolis by two complementary colorimetric methods. J Food Drug
Anal 2002, 10(3):178–182.
26. Fu M, He Z, Zhao Y, Yang J, Mao L: Antioxidant properties and involved Submit your next manuscript to BioMed Central
compounds of daylily flowers in relation to maturity. Food Chem 2009, and take full advantage of:
114(4):1192–1197.
27. Tallarida RJ: Statistical analysis of drug combinations for synergism. Pain
• Convenient online submission
1992, 49(1):93–97.
28. Xia X, Cheng G, Pan Y, Xia Z, Kong L: Behavioral, neurochemical and • Thorough peer review
neuroendocrine effects of the ethanolic extract from Curcuma longa L. in • No space constraints or color figure charges
the mouse forced swimming test. J Ethnopharmacol 2007, 110(2):356–363.
29. Lorke D: A new approach to practical acute toxicity testing. Arch Toxicol • Immediate publication on acceptance
1983, 54(4):275–287. • Inclusion in PubMed, CAS, Scopus and Google Scholar
30. Hao C, Lai W, Ho C, Sheen L: Antidepressant-like effect of lemon essential • Research which is freely available for redistribution
oil is through a modulation in the levels of norepinephrine, dopamine,
and serotonin in mice: Use of the tail suspension test. J Funct Foods 2013,
5(1):370–379. Submit your manuscript at
www.biomedcentral.com/submit